Abstracts

A Blinded MRI Study Of Hemispheric Dysplasia: Identification Of Phenotypes, and Correlation To Presurgical Findings And Postoperative Outcome

Abstract number : PH.09;
Submission category : 9. Surgery
Year : 2007
Submission ID : 8203
Source : www.aesnet.org
Presentation date : 11/30/2007 12:00:00 AM
Published date : Nov 29, 2007, 06:00 AM

Authors :
A. Gupta1, H. Kim1, P. Ruggieri2, A. Chirla1, E. Wyllie1, D. Lachhwani1, P. Kotagal1, W. E. Bingaman1

Rationale: To analyze the dysplastic and contralateral (unaffected) hemisphere on brain MRI in children with hemispheric dysplasia who underwent hemispherectomy. To correlate brain MRI findings to pre-operative clinical and EEG findings, and to post-operative seizure outcome and followup EEG. To study the frequency of contralateral brain MRI abnormalities, and their impact on presurgical evaluation and postoperative outcome. Methods: Of 415 patients who had epilepsy surgery, preoperative brain MRI on 29 (mean surgery age 13 months) with hemispheric dysplasia were systematically analyzed by a blinded neuroradiologist using a predefined criteria applied to the dysplastic and non-dysplastic (contralateral) hemisphere. Preoperative clinical and EEG data, as well as postoperative seizure outcome and EEG follow-up were reviewed. Results: Brain MRI in 29 patients showed 3 groups; involvement of entire hemisphere (12, 41%) or gradient with posterior (12, 41%) or anterior (5, 17%) predominance. Only 9 (31%) patients had typical findings of hemimegalencephaly with dysplasia and overgrowth of the entire hemisphere. Partial overgrowth of dysplastic hemisphere (quadrantic megalencephaly) was noted in patients with anterior (3/5, 60%) and posterior (7/12, 58%) gradient. Polymicrogyria (18/29, 62%) was the most common gray matter abnormality followed by varying combinations of pachygyria, dysmorphic sulcation, and clefts (11/29, 38%). 21 (21/29, 72%) had white matter signal abnormalities with increased volume in 12 (41%). 15 (51%) had lateral ventricular enlargement. Ipsilateral deep gray (3/29, 10%) and cerebellar (2/29, 7%) involvement were rare. Of 29, 19 (65%) had contralateral MRI abnormalities that were subtle; dysmorphic sulcation patterns, decreased volume, or white matter changes. There was no difference in the age of seizure onset (mean 4.3 months), seizure type and frequency, degree of hemiparesis, and cognition in 3 MRI groups. There was no difference in the frequency of preoperative interictal (4/10 vs 12/19) and ictal (8/10 vs 16/18) scalp EEG lateralization in patients with and without contralateral MRI abnormalities. After surgery (mean follow-up 35 months), 22 of 29 (76%) were seizure free. There was no difference in postoperative outcome between patients with and without contralateral MRI abnormalities, and between 3 MRI groups. Follow-up EEG 6 months after surgery showed resolution of contralateral spikes in all seizure free patients irrespective of preoperative contralateral MRI and EEG abnormalities. Conclusions: Hemispheric dysplasia are heterogeneous, and may involve entire hemisphere or occur with a posterior or anterior gradient. Hemimegalencephaly is at the extreme spectrum, and partial overgrowth of hemisphere may occur only posteriorly or anteriorly. Polymicrogyria is most frequent abnormality in the dysplastic hemisphere suggesting possible non-genetic mechanisms. Contralateral MRI abnormalities occur in 65% patients; however, they do not negatively impact the presurgical evaluation, and postoperative seizure outcome.
Surgery