Jaiman Victor Emmanuel MBBS, Presenter: Nothing to Disclose

Charuta Dagia MD, Abstract Co-Author: Nothing to Disclose

Michael Ronald Ditchfield MBBS, Abstract Co-Author: Nothing to Disclose

Lee Torine Coleman MBChB, Abstract Co-Author: Nothing to Disclose

Cathy Catroppa, Abstract Co-Author: Nothing to Disclose

Vicki Anderson, Abstract Co-Author: Nothing to Disclose

This study was undertaken to determine how imaging findings in children with traumatic brain injury evolve and how initial and follow up imaging correlate with functional outcome.

A prospective review of initial (CT) and 5 year follow up (MRI) brain imaging of 30 children (21 male, median age 4 years) who presented with severe head trauma was correlated with neuropsychological outcome at 5 years. The imaging was assessed for lesion number, location, severity and hemorrhage using a modified Coffey Classification. The Vineland Adaptive Behaviour Scale (VABS) was used to assess pre injury abilities and Full Scale Intelligence Quotient (FSIQ) score was used at 5 years.

11 (36.6%) children had normal initial and follow up imaging. Functional assessment was FSIQ score 104 (median 105), mean VABS score pre injury was 103. 9 (30%) children's initial CT was abnormal (small focal regions of parenchymal or extra axial hemorrhage) and follow up MRI normal. Functional assessment was FSIQ score 104 (median 95), mean VABS score pre injury was 103. 3 (10%) children's initial CT was normal and follow up MRI abnormal (confluent regions of frontal and temporal gliosis and encephalomalacia). Functional assessment was FSIQ score 71 (median 70), mean VABS score pre injury was 118. 7 (23.3%) children's initial CT and follow up MRI were abnormal (severe and widespread and most commonly in the frontal and temporal lobes). Functional assessment was FSIQ score 91 (median 95), mean VABS score pre injury was 106. Those with confluent parenchymal lesions on MRI at follow up had FSIQ ≤101. None with FSIQ >102 had a parenchymal lesion. Mean FSIQ of those with no parenchymal lesion was 104 and with a parenchymal lesion 85.

A CT scan at the time of the initial injury was a poor predictor of long term functional outcome. A normal CT did not exclude a severe injury and an abnormal CT scan was not always associated with a poor outcome. The clinical outcome had a high correlation with residual parenchymal changes demonstrated on follow up MRI.

The persistence and extent of parenchymal injury on MR imaging is a useful predictor of functional outcome in children after traumatic brain injury.

Emmanuel, J, Dagia, C, Ditchfield, M, Coleman, L, Catroppa, C, Anderson, V, Traumatic Brain Injury in Children: Correlation of Initial CT, Follow-up MRI, and Functional Outcome.  Radiological Society of North America 2007 Scientific Assembly and Annual Meeting, November 25 - November 30, 2007 ,Chicago IL. http://archive.rsna.org/2007/5009808.html