Comparative performance of head ultrasound and MRI in detecting preterm brain injury and predicting outcomes: A systematic review

Abstract

Aim: To systematically review the literature to compare the performance of head ultrasound (HUS) and magnetic resonance imaging (MRI) in their ability to detect brain injury and their predictive value for neurodevelopmental outcomes.

Methods: This was a systematic review based on literature search in three electronic databases (MEDLINE, EMBASE, Cochrane Library) and additional sources for studies on routine brain injury screening in preterm neonates published during 2000-May 2020. Studies were included if they reported on the comparative performance of HUS and MRI in detecting preterm brain injury and/or their predictive value for neurodevelopmental outcomes. Findings from the included studies underwent narrative synthesis.

Results: Forty-six studies were included. In comparison with HUS, MRI detected more anomalies and provided more details on the severity and the extent of preterm brain injury, particularly for white matter injury and cerebellar haemorrhage. Neonatal neuroimaging predicted outcomes with high negative predictive value but relatively low positive predictive value. The prognostic value of neonatal neuroimaging varied according to several factors including modality and timing of imaging, and tools used for grading brain injury.

Conclusion: Compared with HUS, MRI offers a better characterisation of preterm brain injury and may enhance the ability to predict neurodevelopmental outcomes.

Keywords: MRI; brain injury; cranial ultrasound; neurological prognosis; preterm.