Abstract Archives of the RSNA, 2005
SSQ15-04
Pediatric Neuroimaging of Cortical Dysplasia: MRI Findings and Pathological Correlation
Scientific Papers
Presented on December 1, 2005
Presented as part of SSQ15: Pediatric (Neuroradiology)
Trainee Research Prize - Resident
Carol Chia-chia Wu MD, Presenter: Nothing to Disclose
Joyce Wu MD, Abstract Co-Author: Nothing to Disclose
Dennis J Chute MD, Abstract Co-Author: Nothing to Disclose
Raman Sankar, Abstract Co-Author: Nothing to Disclose
Gary W Mathern MD, Abstract Co-Author: Nothing to Disclose
Noriko Salamon MD, Abstract Co-Author: Nothing to Disclose
To correlate MRI findings with histopathology in pediatric cortical dysplasia (CD) to determine the sensitivity, specificity, and accuracy of qualitative MRI assessments.
A total of 52 operatively proven cases of cortical dysplasia from 1999 to 2004, including 26 males and 26 females, were retrospectively reviewed. Pre-operative MRIs of the brain were performed at mean age of 52.2 months. MRI protocol included axial T2 FSE and T1, sagittal T1, and coronal SPGR sequences. In addition, axial PD sequences were obtained for patients less than 1 year old, and axial FLAIR sequences were obtained for patients more than 1 year of age. MRIs were assessed for 1) gray matter thickness, 2) gray-white matter junction blurring, 3) subcortical T2 signal abnormality, 4) gray matter T2 hyperintensity, and 5) volume loss. Pathological reports were reviewed for the presence of 1) balloon cells, 2) dysmorphic neurons, 3) polymicrogyria, 4) neuronal heterotopia, and 5) dyslaminiation.
Contrary to previous reports, this study demonstrated a relatively poor association between abnormal subcortical T2 hyperintensity and the presence of balloon cells in Taylor type CD(accuracy=60%, p=0.43). However, increased gray matter thickness correlated with the presence of balloon cells (p=0.0118, sensitivity=79%, specificity=51%, accuracy=59%). Abnormal T2W gray matter hyperintensity was also associated with the presence of balloon cells (p=0.0080, sensitivity=80%, specificity=75%, accuracy=75%). Subcortical T2 hyperintensity demonstrated association with the presence of dysmorphic neurons (p=0.0047, sensitivity=67%, specificit=70%,accuracy=69%). Statistical significant associations were also noted between 1) T1 gray-white junction blurring and dysmorphic neurons, 2) gray matter thickening and polymicrogyria, and 3) volume loss and neuronal heterotopia (p<0.05).
Several MRI features demonstrated significant association with histopathological findings of CD with high sensitivity or specificity. However, these association were of only moderate accuracy. Hence, there is considerable room for improvement in MRI assessments of CD in pediatric patients with medically refractory epilepsy.
Wu, C,
Wu, J,
Chute, D,
Sankar, R,
Mathern, G,
Salamon, N,
Pediatric Neuroimaging of Cortical Dysplasia: MRI Findings and Pathological Correlation. Radiological Society of North America 2005 Scientific Assembly and Annual Meeting, November 27 - December 2, 2005 ,Chicago IL.
http://archive.rsna.org/2005/4410463.html