BRAIN MRI CHARACTERISTICS OF EPILEPTOGENIC TUBER AND PREDICTORS OF INTRACTABLE EPILEPSY IN TUBEROUS SCLEROSIS COMPLEX (TSC)
Abstract number :
B.02
Submission category :
4. Clinical Epilepsy
Year :
2009
Submission ID :
10444
Source :
www.aesnet.org
Presentation date :
12/4/2009 12:00:00 AM
Published date :
Aug 26, 2009, 08:12 AM
Authors :
S. Viravan, K. Velayudam, P. Kotagal, D. Lachhwani, E. Wyllie, I. Tuxhorn, W. Bingaman and Ajay Gupta
Rationale: To select TSC patients for epilepsy surgery, epileptogenic tuber(s) (ET) must be identified among many possible candidate tubers. Patients are screened by videoEEG with or without FDG-PET and ictal SPECT, and matched with the inventory of tubers seen on brain MRI in the region of interest. It is not known if ET has unique MRI characteristics that would distinguish it from other tubers on visual analysis alone. Using TSC clinic longitudinal data on 53 TSC patients with epilepsy, we studied predictors of intractability. We also analyzed preoperative MRI of seizure free children to define characteristics of resected ET using other tubers in each patient as controls. Methods: Of 73 TSC patients (Jan 1996-June 2008), longitudinal clinical, EEG and brain MRI data on 53 (27 males) with epilepsy were collected, and variables were compared between nonintractable and intractable (failed 2 drugs, at least 1 seizure/month and no seizure free interval > 3months) groups. 18 children who had epilepsy surgery and presurgical evaluation with a standard protocol brain MRI at the Cleveland Clinic were further studied, and all tubers on brain MRI were analyzed in 10 seizure free patients to determine preoperative MRI characteristics of the resected ET in comparison to nonepileptogenic (nET, remote location and silent on videoEEG), and potentially epileptogenic tuber(s) (pET, in the vicinity of ictal scalp EEG onset but not resected) in each patient. Results: Of 53 patients (age 1-37 years, median 13), 41 (77%) were intractable. Presence of unilateral limb motor seizures had significant association with intractability (p<0.05), while there was a trend to suggest intractability with infantile seizure onset (cut-off < 4 months) and longitudinal presence of multiple seizure types. There was no significant difference between sex, history of infantile spasms, cognitive delay, generalized multiregional vs focal scalp EEG abnormalities, and number and location of tubers on brain MRI between the two groups. Of 18 (age 0.25-18 years, median 6.5) who underwent surgery (12 single tuber, 5 multituber complex, 1 multilobar resection), 10 (56%) were seizure free at 0.5-4 (median 1.5) years follow up. MRI analysis in 10 seizure free patients identified 56 tubers; 15 pET, 16 ET, and 25 nET. Size of mET (28.7+ 13.3mm) was significantly larger than pET (13.7+6.1mm) and nET (14.5+4.1mm) (p < 0.001). Extensive perituberal white matter abnormalities (increased signal in superficial and deep white matter on T2 images) as well as presence of calcification were significantly associated with ET than with pET and nET (p < 0.05). Conclusions: Intractability of epilepsy in TSC is likely in seizures of infantile (<4months) onset and focal motor semiology. Infantile spasms, cognitive delay, brain MRI, and EEG abnormalities are not good predictors of intractability. On brain MRI, epileptogenic tuber(s) are likely to be a single large or a partially confluent multi-tuber complex with extensive deep white matter abnormalities and calcification.
Clinical Epilepsy