The Cleft the Scan Can See, and the One It Usually Cannot

What new research from Croatia, Portugal and the Netherlands tells families about finding a cleft before birth, testing for genetic causes, and deciding when to close the palate

A gap in the roof of the mouth is the hardest kind to spot, and the most likely to affect feeding

A cleft is a gap left where parts of a baby's face did not join during early pregnancy, and it can involve the lip, the roof of the mouth — the palate — or both. Pregnancy scans find a cleft lip easily but rarely find a cleft palate, and the palate is the part that decides whether a newborn can feed. New research measures that mismatch and asks what it should change.

The reason is anatomical. The lip is on the surface of the face and sits in the view sonographers already take; the palate is tucked behind the gum ridge and is far harder to see. Three studies published between September 2025 and August 2026 — from Croatia, Portugal and the Netherlands — have now measured how badly detection and feeding need line up, looked at what genetic testing in pregnancy actually finds, and asked whether operating on the palate earlier helps babies eat better.

How families used to face this, and what changed

For decades, a cleft was something most families discovered in the delivery room. Ultrasound in pregnancy gradually changed that for clefts involving the lip: by the 1990s and 2000s, many were being picked up at the mid-pregnancy anomaly scan, giving families months to meet the cleft team, learn about feeding, and prepare. Researchers then worked hard to push detection earlier still, developing careful first-trimester techniques with names like the "retronasal triangle" [1] and the "maxillary gap" [2] — clues that can suggest a cleft as early as 11 to 13 weeks.

But a review of many studies found something that has never really been solved. When ultrasound was used in ordinary, low-risk pregnancies, it found anywhere from 9% to 100% of clefts involving the lip — but only 0% to 22% of clefts affecting the palate alone [3]. Adding 3D scanning has not reliably changed this [4]. So while families expecting a baby with a cleft lip usually know in advance, families of babies with a cleft palate alone very often do not — and those are the babies most likely to have trouble with their first feeds. Until recently, nobody had measured both of these things in the same group of children to see just how badly the two lined up.

What a study of 328 children in Croatia found

Researchers at Dubrava University Hospital in Zagreb — the single centre where all children in Croatia go for cleft surgery — looked at 328 children who had a cleft and no recognised syndrome [5]. For each child they recorded two things: whether the cleft had been spotted on a pregnancy scan, and how the baby was fed in the first weeks of life.

The pattern ran in exactly opposite directions. Among children with a cleft lip only, 38 in every 100 had been found before birth. Among children with a cleft lip and palate, 34 in 100. Among children with a cleft of the palate alone, just 2 in 100.

Feeding went the other way. Some babies with a cleft need a thin feeding tube passed through the nose into the stomach for a while — a nasogastric tube — because they cannot create the suction that normal sucking requires. Not one of the 74 children with a cleft lip only needed one. Among children with a cleft lip and palate, about 14 in 100 did. Among children with a cleft of the palate alone, about 25 in 100 did — and among those whose palate cleft was complete, nearly 4 in 10.

The single most striking figure is this: of the 46 children who needed tube feeding, 43 — more than 9 in 10 — had not been picked up on a pregnancy scan [5]. The babies most likely to need help eating are almost exactly the babies who arrive without anyone expecting it.

The same study found one other thing worth knowing. Newborns in Croatia have a routine hearing check before going home, and a "refer" result — meaning the test needs repeating — was much more common in babies whose cleft involved the palate (about 26 to 32 in 100) than in those with a cleft lip only (fewer than 3 in 100). A refer result is not the same as hearing loss; it means follow-up testing is needed, and for babies with a palatal cleft, that follow-up should simply be expected.

What genetic testing in pregnancy actually found

A second study, from a university hospital in Porto, Portugal, followed 48 pregnancies in which a cleft had been seen on ultrasound between 2014 and 2023 [6]. In 20 of those 48 (about 4 in 10), the baby had other differences as well — heart, brain, kidney or limb findings. Every single genetic abnormality found in the whole study was in that group: trisomy 13 in eight babies, trisomy 18 in four, and rarer single-gene conditions in three more.

In the other 28 pregnancies, where the cleft appeared to be on its own, every genetic test came back normal. After birth, only two of those 28 children turned out to have anything additional — one an unusual tongue, one a difference in the urinary opening.

This does not mean genetic testing is pointless. It means the question a family is really asking — "is there something else going on?" — is usually answered by the scan itself rather than by the genetic result. The researchers also raise a practical point that families deserve to hear plainly: detailed genetic tests can take weeks, and the answer may not arrive in time to inform decisions that have a legal deadline. Many of the gene changes linked to clefts on their own do not reliably predict what will happen to a particular child, because the same change can affect different people very differently [7]. In families where clefts run in the family, more extensive testing does find causes more often — around 1 in 10 in one study of inherited cases [8] — which is why family history changes the conversation.

Does closing the palate earlier help a baby eat?

Surgery to close the palate is usually done somewhere between 6 and 12 months of age, and teams differ on exactly when. The main reason for choosing earlier is speech. A large international trial called TOPS — for Timing Of Primary Surgery — randomly assigned 558 babies with an isolated cleft palate at 23 hospitals across Europe and South America to surgery at 6 months or at 12 months. By age 5, speech problems caused by air escaping through the nose were less common in the 6-month group: about 9 in 100, compared with 15 in 100 [9].

A third new study asked a different question: does earlier surgery help with chewing? At the Wilhelmina Children's Hospital in Utrecht, the Netherlands, researchers running a trial called CLEFED-1 — short for CLEfts and FEeding Difficulties — randomly assigned babies to have their palate closed at 6–8 months or at 10–12 months, then filmed 60 of them eating at 9, 13 and 17 months. Speech therapists who did not know which operation each child had scored the videos using a standard chewing checklist [10].

Chewing improved steadily with age in both groups, and by 17 months the two groups scored almost identically. There was no chewing advantage to operating earlier. One small detail went the other way: children operated on later were slightly better at moving food side to side with their tongue, possibly because they had more practice steering food around the open palate first. Children in the later group did, however, more often need a second small operation to repair a hole that reopened.

So the honest answer to a parent who asks whether earlier surgery will help their baby eat is: the evidence for earlier closure is about speech, not chewing. The chewing study was also smaller than its own researchers had planned — 60 children instead of 80 — and used a checklist designed for other groups of children [11], so it cannot rule out a small difference.

What this means for your family

If your baby's cleft involves the palate, the most useful thing to take from this research is that a proper feeding assessment in the first days matters — whether or not anything was seen on a scan. Ask for it. Ask who to call if feeds are taking too long, if milk comes back down the nose, or if weight gain stalls after you go home. Needing a feeding tube for a while is common and temporary; it is not a sign that anything has gone wrong.

If a cleft was found on your scan and nothing else was, the reassuring finding is that "nothing else" is usually accurate — in the Portuguese series, more than 9 in 10 apparently isolated clefts really were isolated. And a normal genetic test in that situation is the expected result, not a lucky one. Clefts overwhelmingly happen through a combination of many small genetic and environmental influences rather than a single identifiable cause [12], and researchers are still mapping which factors matter and by how much [13].

What researchers are working on next

Three directions follow from this work. The first is imaging: rather than reporting how many clefts overall are found before birth — a figure dominated by the easy-to-see lip — researchers want to know whether targeted techniques for the palate can move that 2% figure in ordinary care. The second is genetics: larger studies across multiple hospitals are needed before extensive testing in an apparently isolated cleft can be recommended as routine, given how long results take and how hard they can be to interpret. The third is surgery: trials of timing have so far been judged mainly on speech at age 5, and future trials will need to follow feeding, chewing and growth as well, for longer — at least to age four, when chewing has fully matured. Studies that follow the same children from the pregnancy scan to school age are still rare [14], and they are what families will benefit from most.

References

  1. Sepulveda W, Wong AE, Martinez-Ten P, Perez-Pedregosa J. Retronasal triangle: a sonographic landmark for the screening of cleft palate in the first trimester. Ultrasound Obstet Gynecol. 2010;35(1):7-13. doi:10.1002/uog.7484 ↩
  2. Chaoui R, Orosz G, Heling KS, Sarut-Lopez A, Nicolaides KH. Maxillary gap at 11-13 weeks' gestation: marker of cleft lip and palate. Ultrasound Obstet Gynecol. 2015;46(6):665-669. doi:10.1002/uog.15675 ↩
  3. Maarse W, Pistorius LR, Van Eeten WK, et al. Prenatal ultrasound screening for orofacial clefts. Ultrasound Obstet Gynecol. 2011;38(4):434-439. doi:10.1002/uog.8895 ↩
  4. Divya K, Iyapparaja P, Raghavan A, Diwakar MP. Accuracy of Prenatal Ultrasound Scans for Screening Cleft Lip and Palate. J Med Ultrasound. 2022;30(3):169-175. doi:10.4103/jmu.jmu_20_22 ↩
  5. Tarle A, Tarle M, Raguž M, Frković SH, Knežević P. Prenatal Ultrasound Detection and Neonatal Nasogastric Tube Feeding Follow Opposite Gradients Across Orofacial Cleft Phenotypes: A Nationally Ascertained Referral-Centre Cohort Study. Children. 2026;13(9):1124. doi:10.3390/children13091124 ↩
  6. Dias H, Fernandes da Rocha D, Henriques Costa M, et al. Cleft Lip and Palate: Prenatal Diagnosis, Genetic Testing, and Pregnancy Outcomes in a Tertiary Referral Center. Acta Med Port. 2026;39(1):33-40. doi:10.20344/amp.23371 ↩
  7. Kini U. Genetics and orofacial clefts: a clinical perspective. Br Dent J. 2023;234(12):947-952. doi:10.1038/s41415-023-5994-3 ↩
  8. Basha M, Demeer B, Revencu N, et al. Whole exome sequencing identifies mutations in 10% of patients with familial non-syndromic cleft lip and/or palate in genes mutated in well-known syndromes. J Med Genet. 2018;55(7):449-458. doi:10.1136/jmedgenet-2017-105110 ↩
  9. Gamble C, Persson C, Willadsen E, et al. Timing of Primary Surgery for Cleft Palate. N Engl J Med. 2023;389(9):795-807. doi:10.1056/NEJMoa2215162 ↩
  10. Guillaume CHAL, Verhoeven MMR, Eligh AM, et al. Chewing with a cleft palate: a randomized controlled trial. Effect of timing of surgical closure of the palate on mastication in infants with clefts. Clin Oral Investig. 2025;29(9):432. doi:10.1007/s00784-025-06467-2 ↩
  11. Remijn L, Speyer R, Groen BE, van Limbeek J, Nijhuis-van der Sanden MW. Validity and reliability of the Mastication Observation and Evaluation (MOE) instrument. Res Dev Disabil. 2014;35(7):1551-1561. doi:10.1016/j.ridd.2014.03.035 ↩
  12. Babai A, Irving M. Orofacial Clefts: Genetics of Cleft Lip and Palate. Genes. 2023;14(8):1603. doi:10.3390/genes14081603 ↩
  13. Ács M, Cavalcante BGN, Bănărescu M, et al. Maternal factors increase risk of orofacial cleft: a meta-analysis. Sci Rep. 2024;14(1):28104. doi:10.1038/s41598-024-79346-7 ↩
  14. Farladansky-Gershnabel S, Gluska H, Halevi N, et al. Cleft Lip and/or Cleft Palate: Prenatal Accuracy, Postnatal Course, and Long-Term Outcomes. Children. 2022;9(12):1880. doi:10.3390/children9121880 ↩