Which Brain Scan Tells Families More After a Premature Birth: One Taken Soon After Birth, or One Near the Due Date?

A 2026 review of 30 studies (Bonezzi and colleagues) and a reply from brain-imaging experts (Cizmeci, Cowan and de Vries), explained for parents: what each scan can show, and why experts disagree

Babies born early often have a brain scan called MRI (magnetic resonance imaging) in hospital, but doctors disagree about whether a scan a few weeks after birth or one near the original due date better predicts how a child will develop. A 2026 review of 30 studies (2,107 babies) favoured the due-date scan; brain-imaging experts replied that when the comparison was limited to babies with results from both scans, the difference almost vanished. Each scan has its uses, and neither can predict a child's future with certainty.

Why brain scans matter for babies born early

Being born early, or "preterm", means arriving before 37 weeks of pregnancy instead of the usual 40. It is very common: an estimated 13.4 million babies worldwide were born preterm in 2020, about 1 in every 10 births [1]. In one large group of babies born very early, about 1 in 3 showed some sign of brain injury on a scan near their due date, and about 1 in 10 showed serious injury [2]. Many children go on to do well, but some later have difficulties with movement, learning or language.

To understand a baby's brain, doctors may use MRI. An MRI scanner uses a strong magnet and radio waves, not X-rays, to take detailed pictures of the brain. The big question for families and doctors is when to do it: a few weeks after birth, while the baby is still in intensive care, or near the baby's original due date, which doctors call "term-equivalent age".

How doctors came to scan at two different times

About twenty years ago, research teams began scanning the same premature babies more than once. A team in San Francisco scanned 89 babies once they were stable, at around 32 weeks (counting from the start of the pregnancy), and scanned 50 of them again near their due date. Moderate or severe changes on either scan were linked with later developmental problems [3]. A London team that scanned 119 very premature babies several times found that many changes seen early did not clearly predict how the child would do, while some changes seen near the due date did [4]. Researchers in the Netherlands then showed that small spots of injury in the brain's "white matter", its wiring, can look different over time: in about a third of the babies with these spots they were visible only on the early scan, and in a fifth only on the later one [5]. Later, a scoring system was developed so that due-date scans could be read in a consistent way [6], and in 325 very premature babies more serious injury on that scan was linked with slower development and with cerebral palsy, a condition affecting movement and posture [2]. Meanwhile, other researchers combining many studies found that early scans were very good at picking out babies who would later be diagnosed with cerebral palsy [7]. So each scan had supporters, and no one had properly compared them in the same babies.

What the 2026 review did

A team of researchers from Brisbane, Australia, and Florence and Pisa, Italy, led by Bonezzi, set out to fill that gap [8]. They registered their plan publicly before starting, a step that helps keep reviews honest. They searched the medical literature for every study in which premature babies had an early MRI (before 36 weeks) and a due-date MRI (36 to 44 weeks), and in which the children's development was later checked between 1 and 3 years of age with standard developmental tests. They found 30 studies including 2,107 babies. A review like this, which pools the results of many studies, is called a "meta-analysis". Only 9 of the 30 studies reported their results in a way that could be combined.

What the review found

When the researchers looked at each type of scan separately, the due-date scan looked better. Picture 100 children who later had developmental difficulties: an abnormal due-date scan had flagged about 84 of them, and an abnormal early scan about 76. Now picture 100 children who did well: the due-date scan had correctly looked normal in about 73, but the early scan looked normal in only about 45, which means more "false alarms" from early scans [8]. False alarms matter because a worrying scan result can weigh heavily on a family for months, even when the child goes on to do well. A scan that gives fewer false alarms, while still picking up most children who need extra help, is more useful for planning follow-up and for honest conversations with parents. Injury to the white matter seen near the due date was the strongest warning sign for later movement difficulties. Bleeding into the fluid-filled spaces of the brain, called intraventricular haemorrhage, turned out to be a weak predictor of later development at either time.

The researchers concluded that the due-date scan "offers higher predictivity" than the early scan. They also wrote that the two scans play "complementary roles", meaning each adds something different [8].

Why other experts urged caution

About six weeks later, three neonatal brain-imaging specialists from Toronto, London and Bristol, and Leiden wrote a reply in the same journal [9]. They agreed the question was important, but argued that the conclusion was weaker than it looked.

Their main point was about fair comparison. Many of the studies did not report usable results for both scans, so the combined "early" results and the combined "due-date" results came partly from different groups of babies. When the review looked only at the five studies that reported results for both scans in the same babies, the two scans performed almost equally. Of 100 children who later had difficulties, an early scan flagged about 85 and a due-date scan about 81. Of 100 who did well, the early scan looked normal in about 53 and the due-date scan in about 65 [8]. The experts described this difference as "marginal and statistically non-significant", meaning it was small and could have arisen by chance; the review itself did not run a formal test of it.

They raised other concerns too. Some "early" scans, they wrote, were done as late as 35 to 36 weeks, which is not very early at all [9]; the review's own detailed tables suggest that, in the studies that were combined, the typical early scan was at around 31 to 33 weeks, though some babies were scanned at 35 to 36 weeks or even later [8]. Babies stable enough to be scanned early may have been less unwell than those scanned later. And some of the sickest babies never had a due-date scan, because early scan results sometimes help families and doctors make very difficult decisions about care. Finally, development checked at 1 to 3 years cannot show learning difficulties that appear at school age: in one Australian group, children's development at age 2 often did not match how they were doing at 8 [10], and children born early can have effects on learning that last into adulthood [11].

The experts also pointed to research favouring early scans. A Canadian study of 393 very premature babies found that white matter injury seen on the early scan, but not on the due-date scan, was linked with developmental delays at 3 years [12]. Another study following children to age 6 found that early and due-date scans performed broadly similarly for predicting movement problems [13].

What this means for your baby

If your baby is having, or has had, a brain MRI, these findings are reassuring in an important way: both scans give useful information, and doctors choose the timing to suit the question they need to answer. An early scan can be especially helpful after a serious event such as a large brain bleed or a period of low oxygen, when the experts note that some injuries show up most clearly in the first days to weeks [9]. A due-date scan shows how the brain has matured and is often used to plan follow-up. A normal scan at either time is good news, though not a guarantee; an abnormal scan is a reason for closer support, not a final verdict.

Scans are never the whole picture. Doctors combine them with careful examinations of how a baby moves and responds, such as watching a baby's natural movements in the first months, which are among the most useful early signs for cerebral palsy [14]. You can read more about this in our article on spotting cerebral palsy early. Whatever a scan shows, follow-up checks continue, so that any extra help a child needs can be offered as early as possible. Please ask your baby's team what their scan shows, why it was done at that time, and what follow-up is planned.

What researchers are working on next

Between them, the two groups point to what is needed. The review's authors call for scans to be read with the same scoring systems and for children's development to be checked more widely than for cerebral palsy alone [8]. The experts' concerns about scans done at very different ages, and about checking development too soon, point to studies in which every baby is scanned at the same planned ages and children are followed up into their school years [9]. Researchers are also testing whether combining MRI with other tests, such as recordings of brain electrical activity, gives a clearer picture than either alone [15].

The bottom line for families

The 2026 review suggested that a scan near the due date predicts development better than an early scan. A reply from leading experts pointed out that, in the review's own comparison of babies who had results for both scans, the two were nearly equal, and argued that early scans remain important for certain injuries. For families, the message is that both scans can help, neither tells the whole story, and your baby's care team can explain what each one means for your child.

References

  1. Ohuma EO, Moller AB, Bradley E, et al. National, regional, and global estimates of preterm birth in 2020, with trends from 2010: a systematic analysis. Lancet. 2023;402(10409):1261-1271. doi:10.1016/S0140-6736(23)00878-400878-4) ↩
  2. Kidokoro H, Anderson PJ, Doyle LW, Woodward LJ, Neil JJ, Inder TE. Brain injury and altered brain growth in preterm infants: predictors and prognosis. Pediatrics. 2014;134(2):e444-453. doi:10.1542/peds.2013-2336 ↩
  3. Miller SP, Ferriero DM, Leonard C, et al. Early brain injury in premature newborns detected with magnetic resonance imaging is associated with adverse early neurodevelopmental outcome. J Pediatr. 2005;147(5):609-616. doi:10.1016/j.jpeds.2005.06.033 ↩
  4. Dyet LE, Kennea N, Counsell SJ, et al. Natural history of brain lesions in extremely preterm infants studied with serial magnetic resonance imaging from birth and neurodevelopmental assessment. Pediatrics. 2006;118(2):536-548. doi:10.1542/peds.2005-1866 ↩
  5. Kersbergen KJ, Benders MJ, Groenendaal F, et al. Different patterns of punctate white matter lesions in serially scanned preterm infants. PLoS One. 2014;9(10):e108904. doi:10.1371/journal.pone.0108904 ↩
  6. Kidokoro H, Neil JJ, Inder TE. New MR imaging assessment tool to define brain abnormalities in very preterm infants at term. AJNR Am J Neuroradiol. 2013;34(11):2208-2214. doi:10.3174/ajnr.A3521 ↩
  7. George JM, Pannek K, Rose SE, Ware RS, Colditz PB, Boyd RN. Diagnostic accuracy of early magnetic resonance imaging to determine motor outcomes in infants born preterm: a systematic review and meta-analysis. Dev Med Child Neurol. 2018;60(2):134-146. doi:10.1111/dmcn.13611 ↩
  8. Bonezzi L, Rocchitelli L, Biagioni T, et al. Comparative Diagnostic Performance of Early and Term Magnetic Resonance Imaging in Preterm Infants: A Diagnostic Test Accuracy Systematic Review and Bayesian Meta-Analysis. Neonatology. 2026:1-13. doi:10.1159/000552875 ↩
  9. Cizmeci MN, Cowan FM, de Vries LS. Reappraising the Evidence for Term-Equivalent Age MRI Superiority in Preterm Infants. Neonatology. 2026:1-3. doi:10.1159/000553523 ↩
  10. Roberts G, Anderson PJ, Doyle LW. The stability of the diagnosis of developmental disability between ages 2 and 8 in a geographic cohort of very preterm children born in 1997. Arch Dis Child. 2010;95(10):786-790. doi:10.1136/adc.2009.160283 ↩
  11. Hu M, Truong T, Zhu S, et al. Cognitive and Educational Outcomes After Preterm Birth or Low Birth Weight: An Umbrella Review and Meta-Analysis. JAMA Pediatr. 2026;180(6):628-638. doi:10.1001/jamapediatrics.2026.0533 ↩
  12. Selvanathan T, Oloomi S, Guo T, et al. Association of Early-Life or Term-Equivalent White Matter Injury With Neurodevelopmental Outcomes in Very Preterm Infants. Neurology. 2025;105(5):e214016. doi:10.1212/WNL.0000000000214016 ↩
  13. Mistry KH, Bora S, Pannek K, et al. Diagnostic accuracy of neonatal structural MRI scores to predict 6-year motor outcomes of children born very preterm. Neuroimage Clin. 2025;45:103725. doi:10.1016/j.nicl.2024.103725 ↩
  14. Novak I, Morgan C, Adde L, et al. Early, Accurate Diagnosis and Early Intervention in Cerebral Palsy: Advances in Diagnosis and Treatment. JAMA Pediatr. 2017;171(9):897-907. doi:10.1001/jamapediatrics.2017.1689 ↩
  15. Forrest CD, Biagioni T, Liley HG, et al. Combined Use of Electroencephalography and Magnetic Resonance Imaging in the Prognostication of Neurodevelopmental Outcomes in Preterm Infants - A Systematic Review and Meta-Analysis. Pediatr Neurol. 2026;175:116-129. doi:10.1016/j.pediatrneurol.2025.11.005 ↩