J Korean Epilepsy Soc.
2001 Jun;5(1):22-32.
Syndromic Diagnosis at the Epilepsy Clinic: Role of MRI
- Affiliations
-
- 1Department of Neurology, Yonsei University college of Medicine,Seoul, Korea. bilee@yumc.yonsei.ac.kr
- 2Department of Diagnostic Radiology, Yonsei University college of Medicine, Seoul, Korea.
- 3Department of Neurology, Ajou University college of Medicine, Suwon, Korea.
- 4Department of Neurology, Pochon Joongmoon Medicine, Sungnam, Korea.
Abstract
-
PURPOSE: To evaluate the application of MRI to the ILAE classification of epilepsies and epileptic syndromes in the setting of epilepsy clinic.
METHODS
We reviewed epilepsy registry forms, EEG, and MRI of 300 patients who were consecutively registered to the Yonsei Epilepsy Clinic. The algorhithm of syndromic classification consisted of 3 steps ; 1) clinical diagnosis based on the clinical informations described in the registry form, 2) clinical-EEG correlations, and 3) clinical-EEG-MRI correlations. The interictal epileptiform discharges (IEDs) in EEG were divided into focal, multilobar/multifocal, and generalized. MRI-lesions were divided into focal and non-focal (multilobar/multifocal, and diffuse) lesions. The clinical-EEG, EEG-MRI, and MRI-clinical correlations were categorized as concordant, not discordant, and discordant.
RESULTS
Among 300 patients evaluated, 249 patients were found to have epilepsies and both EEG and MRI. By clinical analysis, 190 of 249 patients were diagnosed as localization-related epilepsies (LRE), 24 patients were generalized epilepsies (GE), 34 patients were undetermined epilepsies (UDE), and one patient had alcohol related epilepsy. EEG revealed IEDs in 124 patients and altered the clinical diagnosis in 79 patients. MRI lesions were found in 106 patients with focal lesions in 65 patients and non-focal lesions in 41 patients. MRI lesions were found in 47 of 125 patients with negative EEG. Concordance rates of clinical-EEG, EEG-MRI, and MRI-clinical correlations in 54 patients with lobar epilepsies, who had positive EEG and MRI, were 39%, 54%, and 52%, respectively, and discordant rates were 17%, 11% and 7%, respectively. The complete concordance of all 3 correlations was found in only 33% of them. In 20 patients diagnosed as GE by clinical-EEG correlations, MRI lesions were found in only 3 patients and none of them changed the diagnostic categories due to MRI lesions.
CONCLUSION
In lobar epilepsies, the sensitivity of MRI was quite comparable with EEG and the clinical-MRI correlation was superior to the clinical-EEG correlations. MRI provided additional and complimentary informations and should be incorporated to the ILAE-classification system as the category of 'lesional epilepsy'.
