Background: Preterm babies are at high risk for adverse neurodevelopmental outcomes. Bipolar opinions can still be found on ultrasonography (US) and magnetic resonance imaging (MRI) in detecting brain injury. This study compared the usefulness of both imaging methods and assessed their value in long-term prognosis.
Material/Methods: Twenty-five infants were born between the 22nd and 32nd week of pregnancy: 12 girls, 13 boys. They underwent US examinations (1–12 times) and MRI at term equivalent. MRI was performed with a 1.5 T scanner without sedation. SE/T1-, T2 flair, FSE/T2-, GRE/T2*-weighted images, and DWI were obtained.
Results: MRI provided more details in 6/25 babies. In 1 baby, posterior fossa abnormalities were observed only on MRI. One MRI revealed thin, hypoplastic corpus callosum, while US showed only its anterior part. Cystic white matter lesions were not detected on US in 2 babies, 2 cysts were bilateral on MRI, while unilateral on US. Sonographic diagnosis of 1 cyst was ruled out by MRI. Posthemorrhagic lesions were found on MRI in 3 infants; in 2, they were bilateral; while on US they were unilateral. In 2, cerebellar hemorrhage was visualized only on MRI. In 1 baby, delay in myelination was observed at term equivalent.
Conclusions: MRI at term equivalent proved to be a more sensitive method of detecting hypoxic-ischemic brain injury, cerebral hemorrhage, and congenital anomalies than serial US in preterm infants. Follow-up studies in 2 years will assess the natural history of brain lesions and correlate the neurologic status of children and their brain appearance.

Keywords: ultrasonography (US), Magnetic Resonance Imaging (MRI), Brain, preterm baby