Is sequential cranial ultrasound reliable for detection of white matter injury in very preterm infants?
Related Experiment Video
Updated: Jun 15, 2026
State of the Art Cranial Ultrasound Imaging in Neonates
Published on: February 2, 2015
Lara M Leijser1, Francisca T de Bruïne, Jeroen van der Grond
1Department of Pediatrics, Division of Neonatology, J6-S, Leiden University Medical Center, P.O. Box 9600, 2300 RC, Leiden, The Netherlands.
Insights
Sequential cranial ultrasounds (cUS) can detect severe white matter (WM) injury in preterm infants, but magnetic resonance imaging (MRI) is more reliable for milder cases.
Area of Science:
- Neonatal neurology
- Pediatric neuroimaging
- Developmental neuroscience
Background:
- Cranial ultrasound (cUS) may have limitations in accurately detecting diffuse white matter (WM) injury in neonates.
- Reliable assessment of WM injury is crucial for very preterm infants.
Purpose of the Study:
- To evaluate the reliability of a sequential cUS classification system for WM injury in very preterm infants.
- To compare sequential cUS findings with magnetic resonance imaging (MRI) as the reference standard.
Main Methods:
- 110 very preterm infants (gestational age <32 weeks) underwent serial cUS and MRI around term equivalent age (TEA).
- cUS and MRI were assessed for WM and lateral ventricle abnormalities.
- A classification system (normal/mildly, moderately, severely abnormal) was applied to sequential cUS and MRI data.
Main Results:
- Sequential cUS classified infants as normal/mildly abnormal (22%), moderately abnormal (65%), and severely abnormal (13%).
- MRI classifications were 30%, 52%, and 18% for normal/mildly, moderately, and severely abnormal, respectively.
- The positive predictive value of cUS was high for severe WM injury (0.79) but lower for mild (0.67) and moderate (0.64) abnormalities.
Conclusions:
- Sequential cUS reliably detects severe WM injury in very preterm infants.
- cUS is less reliable for diagnosing mild and moderate WM injury.
- MRI around TEA is recommended for accurate detection of WM injury in this population.
Introduction:
Cranial ultrasound (cUS) may not be reliable for detection of diffuse white matter (WM) injury. Our aim was to assess in very preterm infants the reliability of a classification system for WM injury on sequential cUS throughout the neonatal period, using magnetic resonance imaging (MRI) as reference standard.
Methods:
In 110 very preterm infants (gestational age <32 weeks), serial cUS during admission (median 8, range 4-22) and again around term equivalent age (TEA) and a single MRI around TEA were performed. cUS during admission were assessed for presence of WM changes, and contemporaneous cUS and MRI around TEA additionally for abnormality of lateral ventricles. Sequential cUS (from birth up to TEA) and MRI were classified as normal/mildly abnormal, moderately abnormal, or severely abnormal, based on a combination of findings of the WM and lateral ventricles. Predictive values of the cUS classification were calculated.
Results:
Sequential cUS were classified as normal/mildly abnormal, moderately abnormal, and severely abnormal in, respectively, 22%, 65%, and 13% of infants and MRI in, respectively, 30%, 52%, and 18%. The positive predictive value of the cUS classification for the MRI classification was high for severely abnormal WM (0.79) but lower for normal/mildly abnormal (0.67) and moderately abnormal (0.64) WM.
Conclusion:
Sequential cUS during the neonatal period detects severely abnormal WM in very preterm infants but is less reliable for mildly and moderately abnormal WM. MRI around TEA seems needed to reliably detect WM injury in very preterm infants.