Relevance of brain linear measurements in neonatal care

Background: Preterm birth remains a burden globally and survival has increased. Motor and cognitive impairment are sequelae associated with prematurity and prediction of long-term neurological outcome is important to parents and professionals. The aim of the study was to determine the accuracy of the simple and commonly used method of qualitative cranial ultrasound (CUS) in predicting the future, and then see if simple quantitative measures of brain structure using CUS would improve predictive power. I thus first carried out a meta-analysis of available data to determine the value of qualitative CUS. I then tested the accuracy of CUS measures of the corpus callosum and lateral ventricles, structures that would reasonably be expected to reflect brain connectivity or tissue loss respectively, comparing them to Magnetic Resonance Imaging (MRI) of the same infant. Having defined simple CUS measures that accurately reflected neuroanatomy as defined by MRI, I looked to see if these predicted neurodevelopmental outcome. Methods: Preterm infants born before 33 weeks gestation, raging from 24+0 to 32+6 (mean 30+0), birth weight mean 1.36 kg (range, 0.58 to 2.6) had contemporaneous CUS and MRI performed at a mean postmenstrual age of 42+4 (range, 38+0 to 52+6) weeks. Linear measurements of the corpus callosum and lateral ventricles were compared using reduced major axis regression. 301 infants were included in the study, 11 did not have complete data; therefore 290 infants with CUS/MRI pairs were included in the analysis. Bayley scale of infant and toddler development was performed at a mean corrected age of 20 months (range, 18 to 24). Results: There was a strong linear relationship between the CUS and MRI measurements of the length of the corpus callosum and the lateral ventricles. However these linear measurements were not found to be good discriminators of neurodevelopmental impairment. Conclusion: Although CUS precisely measures the length of the corpus callosum and lateral ventricles; it was not useful at predicting neurodevelopmental outcome at 2 years corrected age.Open Acces