Are Simple Magnetic Resonance Imaging Biomarkers Predictive of Neurodevelopmental Outcome at Two Years in Very Preterm Infants?

Background: Preterm infants are at increased risk of neurodevelopmental impairment due to the vulnerability of the immature brain. Early risk stratification is necessary for predicting outcome in the period of highest neuroplasticity. Several biomarkers in magnetic resonance imaging (MRI) at term equivalent age (TEA) have therefore been suggested.

Objective: To assess the predictive value of simple brain metrics and the total abnormality score (TAS) - a modified score for brain injury and growth - in relation to neurodevelopmental outcome of very preterm infants in MRI at TEA.

Methods: Single-centre cohort study including preterm infants with gestational age (GA) ≤32 weeks and birth weight ≤1,500 g. Biparietal width (BPW), interhemispheric distance, transcerebellar diameter (TCD) and TAS were assessed. To detect subtle haemorrhages, additional susceptibility-weighted imaging (SWI) was used in addition to conventional MRI to evaluate its clinical relevance. Neurodevelopment was tested by the Mental and Psychomotor Developmental Index (MDI/PDI) of the Bayley Scales of Infant Development II at a corrected age of 24 months.

Results: One hundred twenty-nine children with median GA of 28.1 weeks and median birth weight of 980 g were included. BPW significantly correlated with PDI (p= 0.01, R2 = 0.06) and TCD with MDI (p < 0.01, R2 = 0.05) and PDI (p < 0.01, R2 = 0.06) but explained variances were low. TAS was not predictive of neurodevelopmental outcome. By using SWI, additional 4 cases of low grade haemorrhages were identified compared to conventional sequences. In one case this additional information was clinically relevant (MDI/PDI below average).

Conclusion: Simple brain metrics and TAS did not reliably predict neurodevelopmental outcome in a cohort with low prevalence of high grade brain injury. The additional value of SWI is yet to be determined in larger cohorts. The combination of imaging and functional biomarkers may be advisable for the prediction of neurodevelopmental outcome.