Risk Factors for Brain Injury in Newborns Treated with Extracorporeal Membrane Oxygenation.

OBJECTIVE: This study aimed to assess the association of clinical risk factors with severity of magnetic resonance imaging (MRI) brain injury in neonatal extracorporeal membrane oxygenation (ECMO) patients. STUDY DESIGN: This is a single-center retrospective study conducted at an outborn level IV neonatal intensive care unit in a free-standing academic children's hospital. Clinical and MRI data from neonates treated with ECMO between 2005 and 2015 were reviewed. MRI injury was graded by two radiologists according to a modified scoring system that assesses parenchymal injury, extra-axial hemorrhage, and cerebrospinal fluid spaces. MRI severity was classified as none (score = 0), mild/moderate (score = 1-13.5), and severe (score ≥ 14). The relationship between selected risk factors and MRI severity was assessed by Chi-square, analysis of variance, and Kruskal-Wallis tests where appropriate. Combinative predictive ability of significant risk factors was assessed by logistic regression analyses. RESULTS: MRI data were assessed in 81 neonates treated with ECMO. Veno-arterial (VA) patients had more severe injury compared with veno-venous patients. There was a trend toward less severe injury over time. After controlling for covariates, duration of ECMO remained significantly associated with brain injury, and the risk for severe injury was significantly increased in patients on ECMO beyond 210 hours. CONCLUSION: Risk for brain injury is increased with VA ECMO and with longer duration of ECMO. Improvements in care may be leading to decreasing incidence of brain injury in neonatal ECMO patients. KEY POINTS: · Veno-arterial ECMO is associated with more brain injury by MRI compared with veno-venous ECMO.. · Longer duration of ECMO is significantly associated with severe brain injury by MRI.. · Risk for neurologic injury may be decreasing over time with advances in neonatal ECMO..

Regional Brain Biometrics at Term-Equivalent Age and Developmental Outcome in Extremely Low-Birth-Weight Infants.

OBJECTIVE: This study aims to evaluate individual regional brain biometrics and their association with developmental outcome in extremely low-birth-weight (ELBW) infants. STUDY DESIGN: This is a retrospective study evaluating term-equivalent magnetic resonance imaging (TE-MRI) from 27 ELBW infants with known developmental outcomes beyond 12 months corrected age. Regional biometric measurements were performed by a pediatric neuroradiologist blinded to outcome data. Measures included biparietal width, transcerebellar diameter (TCD), deep gray matter area (DGMA), ventricular dilatation, corpus callosum, and interhemispheric distance. The relationship between regional biometrics and Bayley-II developmental scores were evaluated with linear regression models. RESULTS: The study cohort had an average±standard deviation birth weight of 684±150 g, gestational age of 24.6±2 weeks and 48% males. DGMA was significantly associated with both cognitive and motor outcomes. Significant associations were also observed between TCD and corpus callosum splenium with cognitive and motor outcomes, respectively. Other biometric measures were not associated with outcome (p>0.05). DGMA<10.26 cm2 was highly specific for poor motor and cognitive outcome. CONCLUSION: TE-MRI biometrics reflecting impaired deep gray matter, callosal, and cerebellar size is associated with worse early childhood cognitive and motor outcomes. DGMA may be the most robust single biometric measure to predict adverse developmental outcome in preterm survivors.